|
V1KU, evaluation board and
reference design
|
The V1KU board demonstrates the
capabilities of the CogniMem
neural network chip for real-time image recognition. Whether you are
looking for speed performance, accuracy, ease of implementation and
maintenance, V1KU will show you that image recognition can be done by
learning examples and letting its recognition engine handle the
communication between the sensor, the neurons and the outside world. Download
presentation.
|
|
Specifications
- Micron/Aptina
monochrome CMOS sensor (60 fps)
- CogniMem chip with 1024 neurons
(CM1K)
- Actel IGLOO FPGA with 600K
gates programmed with the default CogniSight engine for region
monitoring
- 16 MB of SRAM, 4 MB of Flash memory
- High-speed USB
- RS485 at 921,600 kbits
- 2 opto-isolated output lines
|
Ordering information
Buy now
Other packages
|
|
